Title :
Neural network modeling of dynamical systems
Author :
Bialasiewicz, Jan T. ; Soloway, Don
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Colorado Univ., Denver, CO, USA
Abstract :
A recurrent backpropagation neural network suitable for modelling dynamical systems is analyzed. It is shown that the weight matrices of the neural network model determine, with reasonable accuracy, the impulse response of the modelled dynamical system. By analyzing this impulse response with the eigensystem realization algorithm (ERA), one can obtain state-space representation of the original system. Simulation results are presented
Keywords :
linear systems; neural nets; state-space methods; backpropagation neural network; dynamical systems; eigensystem realization algorithm; neural network model; state-space representation; weight matrices; Feedforward neural networks; Linear systems; Mathematical model; NASA; Neural networks; Neurons; Nonlinear equations; Recurrent neural networks; Sampling methods; State-space methods;
Conference_Titel :
Intelligent Control, 1990. Proceedings., 5th IEEE International Symposium on
Conference_Location :
Philadelphia, PA
Print_ISBN :
0-8186-2108-7
DOI :
10.1109/ISIC.1990.128503
